There is considerable evidence that treatmentfor alcohol disorders can lead to significant improvements in substance use and psychosocial problem severity. However, a significantpercentage of patients relapse to heavy drinking after primary treatment has ended. Patients are therefore frequently referred to continuingcare, or "aftercare," programsin an effort to prevent relapse anddecrease theprobability of additional rehabilitation treatments. However, current models of continuing care may not be adequate for the long-term management of a chronic, relapsing disorder such as alcoholism. One possible approach for improving the long-term management of alcoholism is low intensity monitoring and counseling, with the provision to increase the intensity of treatmentwhen warranted. Alcohol dependent patients who have completed 3 weeks of intensive outpatient treatmentwill be randomly assigned to one of the following interventions: (1) continued participation in standard outpatient treatment withoutadditional intervention (treatment as usual, or TAU); (2)TAU plus brief monitoring and feedback via telephone on a tapered schedule out to 18months (IMF); or (3) TAU plus brief monitoring and counseling via telephone on a tapered schedule out to 18months (TMC). The TMC condition will also include an "adaptive" component, in whichintensity of treatment can be increased when specified criteria have been met The inclusion of the IMF condition will makeit possible to determine whether simple monitoring with minimal feedback, withoutthe addition of counseling and adaptive provisions, is sufficient to maintain good outcomes. Patients will be followed up at 4,8,12,18, and24 months post intake into the study. Follow-up assessments will include measures of drinking and drug use (e.g., self-report, collateral reports, urine, and blood), treatment process and potential mediating factors (e.g., motivation, therapeutic alliance, self-efficacy, mood, social support, self-help involvement), psychosocial problem severity, and utilizationof health and social services. TMF and TMC are predicted to produce better alcohol use outcomes thanTAU. A group by time interaction is also jredicted, in which an effect favoring TMC over TMF will emerge over time. Standard and innovativeanalyses will be done to examine the potential mediating effects of treatmentprocess variables and other factors. Secondary analyses will seek to identify variables that predict which patients benefit to the greatest degree from long-term monitoring. Economic analyses will determine the cost-effectiveness and benefit-cost of TMC and TMF relative to TAU, andto ach other.